Research on early warning of strong earthquake damage in dams based on WPT and SVM
In this study,a new structural damage early warning model is proposed in view of the criticality of dam-age early warning in dam strong earthquake monitoring.The model uses the cumulative energy ratio of wavelet pack-et as a feature extraction tool to accurately capture the dynamic characteristics of the dam structures in strong earth-quakes,and uses the Support Vector Machine(SVM)as an intelligent classifier to analyze the extracted features to achieve accurate early warning of damage state.Through the application of multiple sets of simulation data,the effec-tiveness and high accuracy of the model for complex damage pattern recognition are proved.The results show that the proposed model can provide a more reliable theoretical basis and practical guidance for dam damage warning,and is of great significance for ensuring the safty of dam engineering.
Early warning of dam damageWavelet packet transform(WPT)Cumulative energy ratioSupport vec-tor machine(SVM)
NETL
NSTL
万方数据
